metadata
dataset_info:
features:
- name: label
dtype: int64
- name: input_ids
sequence: int32
splits:
- name: validation
num_bytes: 518000000
num_examples: 500000
download_size: 225368255
dataset_size: 518000000
configs:
- config_name: default
data_files:
- split: validation
path: data/validation-*